Author: Mathias Rothermel


LibTSgm LibTSgm LibTSgm LibTSgm LibTSgm
LibTSgm LibTSgm LibTSgm LibTSgm LibTSgm


LibTSgm is a library for dense multi-view reconstruction. So far it consists of three main modules implementing the functionality for image rectification, dense matching and multi-view structure computation. It provides a C/C++ interface such that the single functionalities can be invoked into your own code. If you are looking for a ready-to-use executable to extract 3D information from your image collection please refer to the SURE homepage. Within the rectification module epipolar images are generated. Input are perspective camera images free of lens distortion and the correspondent interior and exterior orientations. The second module implements SGM-based dense matching. Beside the classical SGM method as published in [2], a more time and memory efficient solution is realized. Within a third module structure is computed from the generated disparity images. It is possible to compute scene structure from multiple disparity maps. In latter case depth maps are fused before the final 3D points or depth images are computed. By exploiting the redundancy of multiple redundant disparity estimations precision of single 3D coordinates is increased and blunders are efficiently filtered. No initial depth information of the scene is required. The modules were tested for different type of sensors and close range-, UAV-, and large frame aerial applications. For detailed information refer to our publication [1]. Check out the gallery to see some of the results. We are grateful for any critics, suggestions and bug reports! If you have generated some nice results we happily add them to our gallery!
Happy matching!





» Windows [64 Bit executables]

» Linux



Here comes pdf document



Version 0.0

  • Hello World!



SURE currently depends on the following third party libraries:


Terms and Conditions

The SURE software package is provided for research purposes only. In case you use this software for publications, please cite reference [1], many thanks! See the LICENSE file provided with the demo software for further details.



[1] Rothermel, M., Wenzel, K., Fritsch, D., Haala, N. (2012). SURE: Photogrammetric Surface Reconstruction from Imagery. Proceedings LC3D Workshop, Berlin ,December 2012

[2] Hirschmüller, H., 2008. Stereo Processing by Semiglobal Matching and Mutual Information. IEEE Transactions on Pattern Analysis and Machine Intelligence 30, 328-341.